Image Credit: MOLPIX/Shutterstock.com
Vision sensor technologies are complex systems that contain many components. Advances in this field are happening all the time. The most popular vision sensors to many will be in smart cameras, which are often used in vehicle recognition technologies. However, vision sensors are also widely used across industrial sectors to monitor processes and ensure product quality. In this article, we look at how these vision sensors are used in industrial applications.
What are Vision Sensors?
Vision sensors have many parts to work and are usually composed of a video camera, display and interface, and a computer processor. Vision sensors require these components to automate industrial processes and decisions, record measurements, initiate pass/fail decisions and observe the quality of analyzable products. Some vision sensors also contain an integral processor – these are called smart cameras.
Vision sensors use images to determine the presence, orientation, and accuracy of surrounding objects. Vision sensors use a combination of both image acquisition and image processing, and multi-point inspections can be done using a single sensor.
Vision sensors also provide a data interchange between the video camera and the computer processing unit. There are two types of vision sensors, the monochrome model, and the color model.
Image Credit: MOLPIX/Shutterstock.com
In the monochrome model, the image is captured by the camera, passes through a lens, and is converted into an electrical signal using a CMOS compatible light-receiving element. The brightness and the shape of the object are then determined by its intensity from each pixel from the light-receiving element. The contrast between the white and black regions can be identified, as can the shape, orientation, and brightness.
The second model is the color model. In this mode, the received light is separated into red, green, and blue colors. This is followed by determining the intensity of each color, even if the colors have minimal intensity differences. The objects are then determined by their color difference rather than their intensity.
Current Uses of Vision Sensors
Vision sensors are widely used throughout industrial plants and industrial processes, to maintain the quality of products and to check that operational efficiency is being achieved. Barcodes, blot/stain detection, sizing and alignment, and many other characteristics can be determined without any contact taking place. Below we look at a few specific ways in which vision sensors are used in engineering and scientific processes.
Contour sensors can recognize and assign previously defined objects. These sensors then inspect a passing object by analyzing its shape and contours against other objects in the process line. Contour sensors can be used to check the structure, orientation, position, and completeness of an object. Contour sensors are often used for quality control, verifying punch-outs in steel rods, verifying automobile wheel nuts, determining the position of scoops in packaging, determining the position of other sensors, and verifying the correct alignment of automobile parts, to name a few.
Pixel Counter Sensor
Pixel counter sensors are used to measure objects by counting the individual pixels of identical gray-scale values within an image. Pixel counter sensors can also group gray-scale values on individual objects to determine their shape, size, and shade. Pixel counter sensors are used to: analyze weld spots, verify missing threads in metal parts, detect registration marks, analyze glue quantity on a joint, detect contrast in assembly processes, verify the correct shape of injection mold products and count the number of holes within a rotor.
Code reader sensors can read barcodes and other 2D codes that read from left to right. These sensors can recognize both the barcode and decode the information within it. These sensors can even read and understand dominos, sudoku puzzles, and different font styles. Applications that utilize this type of vision sensor include reading product packaging labels to see when a product goes off, for sorting products by serial number, and to detect color marks and defect marks, to name a few.
3D sensors scan an object at both its surface and depth. This means that these sensors can be used to analyze the contents inside of the packaging, as well as determining an object by its dimensions. These sensors are used to check if a crate has a full quota of bottles inside it, measuring non-transparent solids and bulk materials in tanks or silos, determining the dimensions of packaging and for volume calculations within warehouses and distribution centers, to name a few examples.
Checking Product Quality
The above sensors and applications are all functions of checking the quality of the product, regardless of the industry in which the product is manufactured. Vision sensors are used in automated processing lines to ensure product quality and uniformity across, and within, batches. Vision sensors are widely used in the food and beverage industry, injection molding assembly lines, the automotive industry, the robotics, and general manufacturing industries. Some sensors are used to inspect the product itself, while others (particularly in the food and beverage industry) focus more on the uniformity and completeness of the packaging.
Sources and Further Reading